Discussion
1.
John Quiggin
I thank the organisers of this conference for inviting me to speak. Before addressing Clifford Winston’s
stimulating paper, I’d like to comment on the main graphic used for this conference, which
shows the Sydney Harbour Bridge. The Bridge is a publicly owned project, constructed under a
fixed-price contract and funded by a toll.
More interesting is the tunnel under the bridge, constructed as one of Australia’s first build, own,
operate and transfer projects. Although the tunnel is notionally private, the revenue flows from the
tunnel are guaranteed by the NSW Government, which is therefore the residual income recipient
and, in economic terms, the owner. The reason for this arrangement was to evade restrictions on
public borrowing imposed at the time by the Loan Council.
The spurious idea that transferring infrastructure to the private sector gives government a ‘magic
pudding’, from which to finance new investment, has been refuted many times, notably by several
other speakers at this conference. Unfortunately, it has been repeated even more often, also
notably by a number of speakers at this conference, including most of those actually involved in
the policy process. The latest variant on this idea, outside the scope of my comments today, is
the idea of ‘capital recycling’.
Coming to Clifford Winston’s paper, it may be divided into two parts: diagnosis and prescription.
The diagnostic element is a critique of US infrastructure policy and the debate surrounding it.
Winston begins by disputing the standard view that investment in infrastructure is inadequate.
Given the inefficiency with which public investment funds are allocated, and with which public
infrastructure is used, Winston argues that it is impossible to determine whether too much or too
little is being spent in aggregate.
As regards investment misallocation, Winston focuses on aspects of the US political process that
lead to projects being allocated on spurious principles of equity, with the idea that each locality
should get a share in every program, regardless of marginal benefit. In addition, there is the
problem of influential politicians securing funds for vanity projects, such as the famous Alaskan
‘bridge to nowhere’. These particular problems are less evident in the Australian system where
political parties play a bigger role than influential individual politicians. However, we have our
own problems, such as the largesse lavished on marginal seats.
Next, Winston considers problems with pricing. Infrastructure services are commonly underpriced
or given away ‘free’, with the implied requirement for rationing. Where prices are charged, they
are neither cost-reflective nor based on the kind of Ramsey pricing principles that would be
recommended by economists.
Although Australia has probably made more progress in this respect than the United States,
neither country has addressed the policy that would probably yield more benefits than any other
188
R e s e rv e b a n k o f Au s t r a l i a
d i s c u s s i on
– that of congestion pricing for roads. It is striking that this issue is regarded as utterly untouchable
by Australian politicians, even politicians who have been willing to commit political suicide in
the pursuit of privatisation. For example, the Bligh Government in Queensland ruled out any
consideration of congestion pricing when the topic was raised in the Henry Tax Review in 2010.
Yet the same government went to a crushing defeat in 2012 on the issue of privatisation – a defeat
predicted by years of devastating opinion polls.
It is true that congestion pricing is a controversial policy, and that the immediate reaction of the
public is generally hostile. But this negative initial reaction does not, based on long experience,
compare with the longstanding and remorseless public opposition to privatisation. In Australia,
polls regularly record majorities of more than 70 per cent opposed to privatisation. It does not
appear that the question has been asked in Australia, but polls taken in the United Kingdom
show majority support for renationalisation of the water and railway industries, which have been
privatised for decades.
The somewhat skewed views of political leaders may be explained by the company they keep. The
small minority of the population who regularly drive into the CBD is heavily over-represented in
the social circles surrounding politicians, as is the even smaller minority who benefit directly from
privatisation as advisers, lobbyists, financiers and so on. So, perhaps it is unsurprising that political
leaders delude themselves into thinking that privatisation is politically saleable and congestion
pricing is not.
Winston’s final concern relates to technology, and the failure of US transport authorities
to embrace advances, some of which are already in use in Australia and elsewhere. Winston
diagnoses ingrained risk aversion on the part of public sector employees, but this seems to be a
cultural problem specific to the US Government.
Although it’s possible to quibble about some of the details of Winston’s diagnosis, few economists
would disagree with its general tenor. Rational investment planning, cost-reflective pricing and
openness to technological innovation all command fairly widespread support.
A decade or two ago, the same might have been said of the prescriptive component of the paper:
privatisation. Even among sceptics, privatisation seemed to be the inevitable wave of the future.
Extensive, and largely disappointing, experience has changed the picture, to the point that
Winston himself is ambivalent. As noted in the paper, ‘[p]olicymakers have privatised infrastructure
in many parts of the world but the preliminary evidence on privatisation’s economic effects is
mixed’ (p 168).
Whereas privatisation in one form or another seemed to be an inevitable trend in the 1990s,
significant countercurrents have emerged in recent years. These include:
••
renationalisation in instances of failed privatisation (Railtrack in the United Kingdom, NZ
railways)
••
new public enterprises undertaking responsibilities previously divested through privatisation
(the National Broadband Network in Australia)
••
reforms to public-private partnership (PPP) systems that reduce risk transfer to the private
sector, and have coincided with a decline in the volume of PPP projects (Australia and
United Kingdom).
C O N F e r e nc e vo l u m e | 2 014
189
disc ussion
What explains the disappointing outcomes of infrastructure privatisation, at least relative to the
high hopes of the late 20th century? Winston’s discussion of the problems of the existing system
gives us a clue. There is a striking mismatch between the problems of the infrastructure sector
and the strengths of private provision, as may be illustrated by Table 1.
Table 1: Infrastructure Policy Needs Versus Private Sector Strengths
Policy needs
Private sector strengths
Cost-reflective pricing
Profit-maximising prices
Efficient allocation of capital
Efficient management of labour
Process innovation
Product innovation
Table 1 may be explicated row by row. As already noted, existing infrastructure systems suffer
from the absence of proper pricing, and from pricing that bears little relationship to the cost of
provision.
It is certainly reasonable to anticipate that privatisation will lead to an increase in prices. However,
the success of private infrastructure investment depends on the ability to find profit-maximising
pricing structures, and these will rarely be cost-reflective. For example, Macquarie Bank, the
leading player in this field, has succeeded largely by virtue of its success in overcoming resistance
to unanticipated price increases, such as the large increase in monopoly parking charges imposed
at Sydney Airport.
As regards operating efficiency, it is commonly argued that public sector infrastructure enterprises
are overstaffed and suffer from restrictive work practices. Private sector operations typically display
more managerial flexibility in these respects. However, labour costs usually make up a relatively
small proportion of the total costs of infrastructure projects. The primary costs are associated
with capital and here the case for privatisation is much less favourable. Not only is equity capital
expensive but the long record of investment bubbles and busts, recently exemplified by the US
dotcom bubble and the global financial crisis, suggest that private sector allocation of capital is
far from optimal.
Finally, as regards innovation, the private sector is clearly superior in identifying and addressing
unmet consumer demands, even demands of which consumers are themselves unaware until new
products and services are offered to them. On the other hand, there is little empirical evidence
to suggest that the private sector does a better job in process innovations like those needed to
improve infrastructure services.
Not only is most fundamental research undertaken within the public sector (broadly defined
to include universities and publicly funded research institutes) but the same is true of critical
technologies. Radar, satellites, GPS, the internet and the World Wide Web, critical to the innovations
discussed by Winston, were all developed in, and first deployed by, the public sector.
To sum up, while the infrastructure sector has plenty of problems, there is no reason to think that
privatisation is likely to provide a solution. Indeed, the observation of Kay and Thompson (1986),
190
R e s e rv e b a n k o f Au s t r a l i a
d i s c u s s i on
nearly 30 years ago, that privatisation is ‘a policy in search of a rationale’ remains just as true today.
I close with their observation that:
the reality behind the apparent multiplicity of objectives is not that the policy [privatisation] has a
rather sophisticated rationale, but rather that it is lacking any clear analysis of purpose or effects;
and hence any objective which seems achievable is seized as justification. (p 19)
Reference
Kay JA and DJ Thompson (1986), ‘Privatisation: A Policy in Search of a Rationale’, The Economic Journal,
96(381), pp 18–32.
2.General Discussion
In response to a question, the discussion began with Clifford Winston commenting on possibly
useful lessons from the US experience of privatisation. Dr Winston described concerns in the
lead-up to the deregulation of US railroads (which are typically monopolies or duopolies) about
potentially large increases in the price of rail services. However, surprisingly, prices of rail services
fell after deregulation. Dr Winston argued that the reason for this was the development of a
contract equilibrium, under which logistics firms would band together and negotiate freight
charges with the railroad, thereby reducing the railroad operator’s relative bargaining power.
Dr Winston went on to suggest that contract negotiations could be useful in ensuring efficient
pricing outcomes from future privatisation. For example, in the case of roads, third parties
representing different subsets of road users (such as car drivers, truck drivers, etc) could negotiate
prices with the road provider. Dr Winston claimed that a system of this kind would achieve welfare
gains, as pricing would respond to the heterogeneous preferences of users. Dr Winston also
commented on shortcomings in the allocation of government funds for infrastructure investment,
arguing that the destination of these funds is often determined by lobbying rather than the
efficient allocation of resources to projects that will have the greatest benefit. He suggested
that privatisation could assist in efficient resource allocation, but went on to stress his belief that
the most constructive way of improving infrastructure provision, specifically for transport, is by
allowing and encouraging innovation by the modes.
Still on the topic of privatisation, one participant asked for Dr Winston’s view on ‘capital recycling’,
which has recently become more popular in Australia. The participant noted that, as well as
providing efficiency benefits, privatisation also provides benefits to the public sector’s balance
sheet by effectively selling a fully operational piece of infrastructure and using the associated
capital to finance further infrastructure investment. Dr Winston agreed that governments would
be attracted to this idea. Another participant suggested that once demand risk has been resolved
(which typically occurs relatively early on in a project’s life), an infrastructure asset behaves like a
government bond, in that it provides a long-term stream of relatively fixed income. The participant
argued that if the shadow cost of government debt is high, it may make sense to sell this income
stream and use the proceeds to pay off the government’s debt obligations.
One participant commented on Dr Winston’s observation that prices for the use of infrastructure
in the United States have tended to be much lower than is economically efficient. The participant
C O N F e r e nc e vo l u m e | 2 014
191
disc ussion
queried the extent to which economically efficient pricing structures could contribute to a closing
of the perceived infrastructure investment gap. Dr Winston opined that an important purpose of
infrastructure pricing is to provide guidance for efficient investment that is based on cost-benefit
analysis. He also suggested that, although the improvement of pricing would do a lot to address
inefficiency problems, a major improvement in pricing structures is unlikely to occur. Another
participant questioned whether the use of high-occupancy-toll (HOT) lanes in the United States
was an efficient way of dealing with the inability to adopt congestion pricing, or whether it
was just another inefficient pricing mechanism. In response, Dr Winston stressed that one of
the key lessons from the US experience of HOT lanes was that it revealed significant preference
heterogeneity among road users – something that government regulation and current operation
of infrastructure does not reveal. However, he was sceptical that efficiency gains will be realised
using this type of pricing. Instead, he argued that improvements in infrastructure will be realised
by modal innovations, specifically through automation (such as driverless cars). Dr Winston also
posited that autonomous vehicles will be invaluable in dealing with ageing population problems,
as the elderly will want to continue driving but may pose a danger to themselves and other
road users.
192
R e s e rv e b a n k o f Au s t r a l i a